1. Field of the Invention
The invention concerns a method to automatically generate a selective MR image as well as a correspondingly designed magnetic resonance system, a corresponding computer program product and a corresponding electronically readable data medium.
2. Description of the Prior Art
Imaging by means of nuclear magnetic resonance, i.e. magnetic resonance tomography or MR tomography, has found an ever broader field of application in medical diagnostics.
Magnetic resonance technology (in the following the abbreviation MR stands for magnetic resonance) is a known technique with which images of the inside of an examination subject can be generated. In simplified form, the examination subject is positioned in a comparably strong, static, homogeneous basic magnetic field (field strengths of 0.2 Tesla to 7 Tesla or more) in an MR apparatus so that nuclear spins in the subject orient along the basic magnetic field. For spatial coding of the measurement data, rapidly switched gradient fields are superimposed on the basic magnetic field.
To trigger nuclear magnetic resonance signals, radio-frequency excitation pulses are radiated into the examination subject, the triggered nuclear magnetic resonance signals are measured and are stored as raw data in k-space, on the basis of which raw data MR images are reconstructed. MR imaging enables image contrasts that result from the combination of multiple parameters. Important MR parameters are, for example, the density of the excited nuclear spins (primarily hydrogen protons), the relaxation times for magnetizations (T1, T2, T2*) of the examined tissue, the magnetization transfer, and diverse other contrast mechanisms.
Magnetic resonance tomography lends itself to new fields of use through the acquisition of MR data with very short echo times TE (for example TE<500 μs), wherein the echo time corresponds to the time period between the excitation of the nuclear spins and the measurement of the nuclear magnetic resonance that is thus triggered. It is thereby possible to show substances or tissue that cannot be depicted by means of conventional sequences for example a (T)SE (“(Turbo) Spin Echo”) sequence or a GRE (“Gradient Echo”) sequence, since their T2 time (the relaxation of the transverse magnetization of the substance or tissue) is markedly shorter than the echo time, and thus a corresponding signal from these substances or tissues has already decayed at the point in time of acquisition. With echo times that lie in the range of the corresponding decay time, it is possible, for example, to show bones, teeth or ice in an MR image even though the T2 time of these objects lies in a range from 30-80 μs.
According to the prior art, sequences are known that enable a very short echo time. One example is the radial UTE (“Ultrashort Echo Time”) sequence as described in, for example, the article by Sonia Nielles-Vallespin “3D radial projection technique with ultrashort echo times for sodium MRI: Clinical applications in human brain and skeletal muscle”, Magn. Res. Med. 2007; 57; P. 74-81. In this sequence type the gradients are ramped up after a wait time T_delay after a non-selective or slice-selective excitation, and the data acquisition is begun at the same time. The k-space trajectory scanned in such a manner after an excitation proceeds radially outwardly from the k-space center. Therefore, before the reconstruction (by means of Fourier reconstruction) of the image data from the raw data acquired in k-space, the raw data must first be converted into a Cartesian k-space grid (for example by regridding).
An additional approach in order to enable short echo times is to scan k-space in points with the free induction decay (FID) signal being detected. Such a method is also designated as a single point imaging because essentially only one raw data point in k-space is acquired per RF excitation. One example of such a method for single point imaging is the RASP method (“Rapid Single Point (RASP) Imaging”, O. Heid, M. Deimling, SMR, 3rd Annual Meeting, Page 684, 1995). According to the RASP method, one raw data point in k-space, the phase of which was coded by gradients, is read out at a fixed point in time after the RF excitation at the “echo time” TE. The gradients are modified by means of the magnetic resonance system for each raw data point or measurement point, and thus k-space is scanned (filled) point by point as is shown in FIGS. 1a and 1b. 
There are many applications of magnetic resonance tomography in which it is desired to differentiate different tissue types.
For example, in the case of tissue types with different chemical shifts, a different magnetic field results at the nucleus, which leads to different resonance frequencies. In the signal acquisition this leads to different phase angles of the two components. The most prominent representatives of two different tissue types in the magnetic resonance signal are fat and water, but other applications are also possible. The resonance frequencies of fat and water differ by approximately 3.3 ppm (parts per million). One method for separation of the signals of two different tissue types (for example fat and water) is the utilization of the phase information of acquired MR signals.
Furthermore, there is the possibility to differentiate various tissue types based on their different time constants, for example T2 or T2*. For this purpose, it is known to acquire two MR images such that the first MR image corresponds to raw data which were acquired at a first echo time TE1 after the at least one excitation pulse of the imaging sequence and that the second MR image corresponds to raw data which were acquired at a second echo time TE2 (with TE1≠TE2, for example TE1<TE2) after the same excitation pulse or, respectively, the same excitation pulses of the imaging sequence. Each of the MR images includes signals of tissues with a time constant of the decay of the transversal magnetization (T2) for which it applies: T2 (tissue)≧TEi (i=1 or 2).
Two MR images are thus acquired, wherein the MR image that corresponds to raw data which were acquired at the echo time TE1 (given TE1<TE2) can include signals of more tissues than the MR image that corresponds to raw data which were acquired at the echo time TE2 (since TE1<TE2≦T2 (tissue)). For example, by pixel-by-pixel subtraction of the two MR images from one another, the tissue that is contained only in the MR image with the shorter echo time can be shown separately or masked out, and thus the tissue types can be selectively displayed.
However, it should to be noted that the two MR images have different intensities (signal strengths) due to the different echo times, depending on the T2 values of the imaged tissue. Therefore, before a subtraction of the MR images from one another, it is necessary to compensate for these intensity differences (for example by weighting factors) in order to be able to actually erase the signals of the unwanted tissue.